Correlation between drug sensitivity determined by clonogenic cell assay and clinical effect of chemotherapy in patients with primary lung cancer.

Correlation studies between the sensitivity of tumor cells determined by clonogenic cell assay and the clinical effect of chemotherapy were performed in patients with primary lung cancer. Thirty-four of 48 patients showed adequate colony formation (greater than or equal to 30 colonies) to test in vitro chemosensitivity. Colony formation of tumor specimens did not differ with regards to prognostic factors such as sex, age, stage, performance status, and prior chemotherapy. The two criteria of in vitro chemosensitivity employed were greater than 70% and 50% reduction in colony numbers. When in vitro chemosensitivity was defined as a colony reduction of greater than 50%, the true positive rate in the assay was 57%, while the true negative rate was 85%. In this case, the predictive accuracy was 79%. When in vitro chemosensitivity was defined as a colony reduction of greater than 70%, the true positive rate for the assay was 100%, while the true negative rate was 81%. In this case, the predictive accuracy was 82%. By Fisher's exact probability test, the overall in vitro/in vivo correlation was statistically significant (P = 0.042 less than 0.05) with the 50% cut-off point, but was not significant (P = 0.053 greater than 0.05) with the 70% cut-off point.